Role of MDA5 and interferon-I in dendritic cells for T cell expansion by anti-tumor peptide vaccines in mice

Abstract

Cytotoxic T lymphocytes (CTLs) are effective components of the immune system capable of destroying tumor cells. Generation of CTLs using peptide vaccines is a practical approach to treat cancer. We have previously described a peptide vaccination strategy that generates vast numbers of endogenous tumor-reactive CTLs after two sequential immunizations (prime-boost) using poly-ICLC adjuvant, which stimulates endosomal toll-like receptor 3 (TLR3) and cytoplasmic melanoma differentiation antigen 5 (MDA5). Dendritic cells (DCs) play an important role not only in antigen presentation but are critical in generating costimulatory cytokines that promote CTL expansion. Poly-ICLC was shown to be more effective than poly-IC in generating type-I interferon (IFN-I) in various DC subsets, through its enhanced ability to escape the endosomal compartment and stimulate MDA5. In our system, IFN-I did not directly function as a T cell costimulatory cytokine, but enhanced CTL expansion through the induction of IL15. With palmitoylated peptide vaccines, CD8α+ DCs were essential for peptide crosspresentation. For vaccine boosts, non-professional antigen-presenting cells were able to present minimal epitope peptides, but DCs were still required for CTL expansions through the production of IFN-I mediated by poly-ICLC. Overall, these results clarify the roles of DCs, TLR3, MDA5, IFN-I and IL15 in the generation of vast and effective antitumor CTL responses using peptide and poly-IC vaccines.

Keywords: Dendritic cells; Interleukin-15; MDA5; Peptide vaccines; Poly-IC; Type-I interferon.

Conflict of interest statement

Conflict of interest

A. Salazar is President and CEO of Oncovir, Inc. and is developing poly-ICLC (Hiltonol ™) for the clinic. Esteban Celis has filed patent applications based on the use of synthetic peptides and poly-IC combinatorial vaccines. The rights of the patent applications have been transferred to the Moffitt Cancer Center (Tampa, FL). Other authors declare no conflict of interest.

Figures

Fig. 1. Role of DCs for T cell expansion induced by peptide vaccination

a, CD11cDTR BM chimeric mice received 105 naïve OT-I cells followed a day later by prime-boost vaccinations with pam-Ova (pam2-KMVESIINFEKL) and poly-ICLC administered 14 days apart. DCs were depleted with diphtheria toxin (± DT) injections prior to the prime, boost or both as indicated. Numbers of OT-I cells in the spleens were evaluated 7 days after the boost. b, CD11cDTR BM chimeric mice received 105 naïve OT-I cells, were primed with pam-Ova + poly-ICLC and were boosted with pam-Ova or mini-Ova (SIINFEKL) + poly-ICLC. Numbers of OT-I cells in the spleens were determined 7 days after the boost. c Mixed BM chimeric (CD11cDTR and b2M-KO) mice were treated and evaluated as described in b. d, CD11cDTR BM chimeric mice were treated and evaluated as described in b with the addition of twice daily injections of IFNb or one dose of 100 μg aCD40 mAb. e, CD11cDTR BM chimeric mice were treated and evaluated as described in b with boosters using mini-Ova with poly-ICIC or poly-ICPEI or using pam-Ova with poly-ICPEI. f, WT mice received 105 naïve OT-I cells followed by a pam-Ova + poly-ICLC prime and 14 days later were boosted with mini-Ova alone, mini-Ova + 2 daily injections of IFNb or mini-Ova+ poly-ICLC. Results are presented as individual mice (each symbol) with the mean ± SD for each group.

Fig. 2. Role of DC subsets in responses to poly-ICLC and pam-peptide crosspresentation

a–b, WT or BATF3-KO mice were vaccinated with pam-Trp1 (pam2-KMVFTAPDNLGYM) + poly-ICLC and the Trp1-specific CTL percentages in blood (a) and total numbers in spleens (b) were evaluated. c, WT or BATF3-KO mice received 105 naïve OT-I cells followed by one dose of pam-Ova + poly-ICLC and 7 days later the absolute numbers of OT-I cells were evaluated in spleens. d–f, WT or BATF3-KO mice were injected with poly-ICLC (50 μg, i.v.) and 18 h later the serum levels of IFN-I, IL15/IL15Ra complexes, and IL12p40 were measured. Results are presented as individual mice (each symbol) with the mean ± SD for each group.

Fig. 3. Differences in poly-IC formulations affect CTL responses

a, WT mice received 105 naïve OT-I cells followed by prime-boost vaccination (administered 14 days apart) with pam-Ova with poly-ICLC or poly-IC and the OT-I cells numbers in spleens were measured 7 days after the boost. b, Similar experiment as in a, except that endogenous CTL responses to the Trp1 epitope were evaluated. c, WT, TLR3-KO or MDA5-KO mice were vaccinated and evaluated as described in a. d, WT and MDA5-KO mice were inoculated s.c. with 3×105 B16F10 melanoma cells followed by ACT of 105 naïve TgTR1 CTLs on day 6 after tumor inoculation. Mice were vaccinated on days 7 and 18 with pam-Trp1 using either poly-ICLC or poly-IC. Tumor growth was monitored over time measuring two opposing diameters. Results are presented for individual mice (each symbol) with the mean ± SD for each group (a–c) or mean tumor size for 5 mice/group with SD (d).

Fig. 4. Poly-ICLC stimulates MDA5 more effectively than poly-IC through the proton sponge effect

a–b, The CD8aDC cell line was stimulated with various doses of poly-IC or poly-ICLC and 18 h later the levels of IFN-I and IL12p40 were measured in the supernatants. c–d, BM DCs from WT, TLR3-KO, MDA5-KO or TLR3/MDA5 KO mice were stimulated with 10 μg poly-IC or poly-ICLC and 18 h later the levels of IFN-I and IL12p40 were measured in the supernatants. e–f, WT mice were injected (i.v.) with 50 μg poly-ICLC or poly-IC and 18 h later the amounts of IFN-I and IL12p40 were measured in serum. g–h, WT, TLR3-KO, MDA5-KO or TLR3/MDA5-KO were injected (i.v.) with 50 μg poly-ICLC and 18 h later the amounts of IFN-I and IL12p40 were measured in serum. i, The CD8aDC cell line was incubated with bafilomycin A1 6 h before stimulation with 50 μg/ml poly-ICLC and the production of IFN-I and IL12p40 was measured. Results are presented for individual replicate samples/mice (each symbol) with the mean ± SD for each group.

Fig. 5. Role of IFN-I in the T cell expansion mediated by peptide and poly-ICLC

a, WT and IFNabR-KO mice received equal numbers (5 × 104) of IFNabR-KO OT-I cells (CD45.1+/CD45.2+) and WT OT-I cells (CD45.1+/CD45.2-) followed by pam-Ova + poly-ICLC vaccination. The percentages of OT-I cells were evaluated in blood 7 days after prime and boost and the percentages of IFNabR-KO OT-I and WT OT-I cells after the boost are shown. b, Total numbers of the OT-I populations in spleens from the experiment shown in a. c, Cytokine production in OT-I cells from a stimulated with 10 μg/ml mini-Ova peptide evaluated by intracellular cytokine staining. d, WT mice were inoculated s.c. with 3 × 105 B16F10-Ova cells and after 6 days received ACT of 105 IFNabR-KO OT-I or WT OT-I cells, followed one day later by pam-Ova + poly-ICLC prime and a boost on day 18. Results are presented for individual replicate samples (b–c) or mean tumor size for 5 mice/group with SD (d).

Fig. 6. IFN-I signaling leads to IL15 production enabling CTL expansion

a, WT or IFNabR-KO mice received 105 naïve IFNabR-KO OT-I cells followed a day later with a pam-Ova + poly-ICLC vaccine prime. Fourteen days later mice received pam-Ova + poly-ICLC or mini-Ova + poly-ICLC boosts, and the total numbers of OT-I cells in spleens were evaluated 7 days later. b, WT BM into IFNabR-KO or IFNabR-KO BM into WT chimeric mice received 105 naïve IFNabR-KO OT-I cells followed by a pam-Ova + poly-ICLC prime followed 14 days later by mini-Ova boost administered alone (−) or in combination with poly-ICLC or IFNb. Numbers of OT-I cells in spleens were evaluated 7 days after the boost. c, WT mice received 105 naïve OT-I cells followed by 2 doses of pam-Ova + poly-ICLC vaccine 14 days apart. Some mice received aIL15 mAb on days 14, 16 and 18 and the numbers of OT-I cells in spleens were evaluated 7 days after the boost. d, Similar experiment as in c, except that IFNabR-KO OT-I cells were transferred into IFNabR-KO mice. e, WT or IFNabR-KO mice were vaccinated with 2 doses of pam-Trp1 + poly-ICLC 14 days apart with or without aIL15 mAb treatment on days 14, 16 and 18 and the endogenous responses to the Trp1 epitope were evaluated 7 days post-boost. f, IFNabR-KO mice received 105 naïve IFNabR-KO OT-I cells followed by pam-Ova + poly-ICLC vaccine prime followed 14 days later with pam-Ova + poly-ICLC or mini-Ova + poly-ICLC boosts alone or in combination with 100 μg of aCD40 mAb or IL2Cx (on days 14, 16 and 18). Numbers of OT-I cells were evaluated in spleens 7 days after the boost. Results are presented as individual mice (each symbol) with the mean ± SD for each group.